Hi Andreas and everybody else,
On 20.03.25 11:19, Andreas Tille wrote:
2. The source package urjtag is maintained in Debian Electronics team.
The package was not updated since 2016 but Git received a major update
by
Christian Kreidl<deb...@chk.cksf.de> Wed, 31 Jan 2024 11:58:27 +0100 [2]
It was updating the code to the latest upstream.
the latest upstream development for urjtag was done in 2021. The latest
commit for liburjtag was done in 2024.
So you hijacked binary packages and replaced them with a much older version.
3. I did a team upload of this code base after upgrading the packaging at
Fri, 21 Feb 2025 19:10:19 +0000
which went to NEW since the packaging attempt by Christian did split out
liburjtag0 and liburjtag-dev. I was discussing this upload long
ago (Wed Oct 16 19:42:45 BST 2024 on pkg-electronics mailing list[10])
Oh come on, why do you call your email from Oct 2024 a discussion? It
was just one email and an answer that Geert is not really interested in
this package anymore.
6. I analysed the situation and kept all involved parties in CC at the
next day[5]. I got no answer.
You explicitly addressed the electronics team and asked for an evaluation.
As I now understand nobody in this team was interested in that package
anymore, so nobody was going to answer your request.
I did answer your second email and asked you to clean up your mess.
8. Thorsten answered[7] only to the list which I do not read regularly
in a way that I was asking a LLM for characterisation of Thorsten's
response:
I had hoped that you revert your mistake and clean up the mess on your own.
The wording came across as 'critical and somewhat confrontational,'
'blunt,' 'disapproving,' and 'uncooperative.' Because of that, it's
difficult for me to assume good intentions when receiving a message
like this-but I'm trying anyway. Thorsten, I'd appreciate it if you
could be more mindful of your wording in the future. Thank you.
You did a mistake and even admitted in [5] that you did something wrong.
But instead of fixing your error, you declared that everything is fine.
So yes I was angry about your behaviour and my email put this into words.
If I understand you correctly, you think that by doing the autopkgtest
of libahp-xc with your old urjtag packages, everything is fine now?
So all the new development in liburjtag is just a joke and not needed?
Did you even have a look at the tests of libahp-xc to see whether your
statement is only roughly true?
Did you understand that the purpose of liburjtag is to communicate with
some hardware? How do you think a software autopkgtest can verify
whether this communication is working?
Did you check whether the AHP XC Crosscorrelators, which libahp-xc
supports, are still accessible with your version of liburjtag?
9. Thorsten uploaded liburjtag[8] 2024.03.24-1 on
Wed, 05 Mar 2025 19:06:36 +0100
with no notification of any involved party.
You are blaming me for not notifying anybody but your uncoordinated
upload a few days before is totally fine?
I don't agree that this is simply a "mess that I need to clean up
alone." I asked for technical advice to find the best solution and
offered help in resolving it. Thorsten, if you have specific concerns,
I'd prefer to discuss them constructively.
You did ask after the harm was done.
Your way of communicating is not helpful in solving the problem. In your
mail [7], you neither explained what exactly is broken nor provided any
hints on what might be a good solution.
I wonder why I do have to explain to you why your hijack was wrong.
Shouldn't you tell me whether urjtag is able to be a drop in replacement
for liburjtag?
>From my perspective, renaming one of the liburjtag binary packages would
be a reasonable approach. While the first-come, first-served rule
usually applies, I believe it would make sense for the main urjtag
project to provide the appropriately named libraries, whereas a fork
should have a distinct name.
Until your upload urjtag never provided any binary liburjtag* packages,
so all reverse dependencies do not need them.
Everything was fine before your rash upload, so why do you think that
any change could improve the situation?
The only thing you need to do now is to fix urjtag again.
Are you still considering your behavior as acceptable for a Debian
Developer?
Thorsten
[5]https://alioth-lists.debian.net/pipermail/pkg-electronics-devel/2025-February/012292.html
